Computational Science Technical Note CSTN-143

CSTN Home

Simulating Species Interactions and Complex Emergence in Multiple Flocks of Boids with GPUs

A. V. Husselmann and K. A. Hawick

Archived August 2011

Abstract

The Boids model of spatial agents has proved a great source of insights into how localised microscopic agent-agent interactions can lead to emergent complexity at a macroscopic spatial scale. We explore the interactions between multiple separate species of boids where the flocks maintain a level of coherence amongst their own species but exhibit a preference to separate from other species. We have used data parallelism and Graphical Processing Units(GPUs) to explore large scale flock sizes in interactive simulation time. We also describe some space-covering histogram measurements and cluster component labelling analysis that characterise the flock separation emergent behaviour and scaling in terms of the species interaction parameters.

Keywords: boids; multi-species; flocking; agent-based modelling; data parallelism; GPU; component labelling; emergence; spatial agents; complexity.

Full Document Text: PDF version.

Citation Information: BiBTeX database for CSTN Notes.

BiBTeX reference:

@INPROCEEDINGS{CSTN-143,
  author = {A.V. Husselmann and K.A. Hawick},
  title = {Simulating Species Interactions and Complex Emergence in Multiple
	Flocks of Boids with GPUs},
  booktitle = {Proc. IASTED International Conference on Parallel and Distributed
	Computing and Systems (PDCS 2011)},
  year = {2011},
  editor = {{T. Gonzalez}},
  pages = {100-107},
  address = {Dallas, USA},
  month = {14-16 Dec},
  publisher = {IASTED},
  doi = {10.2316/P.2011.757-012},
  timestamp = {2012.03.16}
}


[ CSTN Index | CSTN BiBTeX ]